Many integrated circuits make use of what is known as a buried contact. A buried contact is an electrical contact formed by the physical contact of a polysilicon layer (a layer which is at least substantially polycrystalline silicon) to a doped region in a semiconductor substrate. This type of contact is most common in MOS type circuits. The desirability of being able to selectively make electrical contact between a polysilicon layer and a doped region in the substrate is well known. This ability is critical, for example, to most static random access memories (SRAMs).
One of the characteristics of buried contacts, the physical contact between a polysilicon layer and the substrate creates a problem in etching. Typically, the substrate is exposed and a polysilicon layer is deposited in physical contact with the substrate. The polysilicon layer is then etched to leave polysilicon only where it is needed. This etching occurs in part on a portion of the area where there is physical contact between polysilicon and substrate which is also typically silicon. This etch process includes some amount of over-etch to ensure that all of the polysilicon is removed where it is intended to be removed. This means that the etch process will continue after the polysilicon in most areas has been removed. After the polysilicon has been removed in the location which is in contact with the substrate, the substrate is etched. A typical amount of over-etch is 20%. Thus, for a 4000 Angstrom polysilicon layer, there will be 800 Angstroms etched out of the substrate because it, being silicon, is etched at virtually the same rate as the polysilicon layer. This over-etching and the consequent etching into the substrate, called pitting, tends to result in defects at those pitted locations in the substrate.